Webscraping-Script.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Webscraping and beyond"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Ruthger Righart**\n",
    "\n",
    "**Email**: rrighart@googlemail.com\n",
    "\n",
    "**Website**: www.rrighart.com\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data is the new oil of the 21st century<font color=\"gray\"><sup>1</sup></font>. Many data come from the web<font color=\"gray\"><sup>2</sup></font>. Acquiring these webdata and structuring them is an essential skill that is called *webscraping*.\n",
    "\n",
    "In this blog, I will briefly discuss how to webscrape data. However, it does not stop there. Analyses beyond webscraping are often needed. Some additional steps are shown, such as capturing hidden characters, merging different data, summary stats and visualization. Luckily, Python can be used for the whole analysis pipeline.\n",
    "\n",
    "Healthdata are used for the current purpose, but it should be mentioned that many data can be webscraped in a similar way, sometimes with a few adaptations in the code.\n",
    "\n",
    "**Important note**: Almost nothing is more dynamic than webdata. Please report me if code is not working, this may be due to changes in the downloaded webdata tables.\n",
    "\n",
    "The current blog has been updated at **21-9-2018**. A previous version was published June 2017 at https://rrighart.github.io/ . The code is based on **Python 2.7**. Some minor changes may be needed for Python 3.4 and later versions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Healthcare rankings for different European countries** "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "**Beautiful Soup** is a Python package that is used for webscraping<font color=\"gray\"><sup>3</sup></font>. \n",
    "In this blog we are going to scrape tables from webpages."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from bs4 import BeautifulSoup\n",
    "import requests\n",
    "import csv\n",
    "import re\n",
    "import urllib2\n",
    "from datetime import datetime\n",
    "import os\n",
    "import sys\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, you need to determine the directory where you save your data.  "
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "os.chdir('/your-directory/')"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "os.getcwd()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are going to scrape data from wikipedia<font color=\"gray\"><sup>4</sup></font>. The data indicate rankings on different health indices, such as patient rights and information, accessibility (waiting time for treatment), outcomes, range and reach of services provided, prevention, and pharmaceuticals. The data are from the Euro Health Consumer index. In the following code, we read the data and use BeautifulSoup to convert the data in so-called **bs4.BeautifulSoup** data. Several data visualizations using Tableau can be found on my blog page<font color=\"gray\"><sup>5</sup></font>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bs4.BeautifulSoup"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "url = 'https://en.wikipedia.org/wiki/Healthcare_in_Europe' \n",
    "r = requests.get(url)\n",
    "HCE = BeautifulSoup(r.content)\n",
    "type(HCE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We could display the data using <font color=\"blue\">print(HCE.prettify())</font>. We will not do that here because it will produce a great amount of text (or soup)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We first need to select the table that we'd like to scrape. As many webpages contain multiple tables, it would be good to read from the HTML the specific tablenames into a list, that we will call <font color=\"red\">lst</font><font color=\"gray\"><sup>6</sup></font>:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "htmlpage = urllib2.urlopen(url)\n",
    "lst = []\n",
    "for line in htmlpage:\n",
    "    line = line.rstrip()\n",
    "    if re.search('table class', line) :\n",
    "        lst.append(line)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This list <font color=\"red\">lst</font> has a length of 3:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(lst)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let us display <font color=\"red\">lst</font>:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['<table class=\"wikitable sortable\">',\n",
       " '<table class=\"wikitable sortable\">',\n",
       " '<div role=\"navigation\" class=\"navbox\" aria-labelledby=\"Healthcare_in_Europe\" style=\"padding:3px\"><table class=\"nowraplinks collapsible autocollapse navbox-inner\" style=\"border-spacing:0;background:transparent;color:inherit\"><tbody><tr><th scope=\"col\" class=\"navbox-title\" colspan=\"2\"><div class=\"plainlinks hlist navbar mini\"><ul><li class=\"nv-view\"><a href=\"/wiki/Template:Healthcare_in_Europe\" title=\"Template:Healthcare in Europe\"><abbr title=\"View this template\" style=\";;background:none transparent;border:none;-moz-box-shadow:none;-webkit-box-shadow:none;box-shadow:none; padding:0;\">v</abbr></a></li><li class=\"nv-talk\"><a href=\"/wiki/Template_talk:Healthcare_in_Europe\" title=\"Template talk:Healthcare in Europe\"><abbr title=\"Discuss this template\" style=\";;background:none transparent;border:none;-moz-box-shadow:none;-webkit-box-shadow:none;box-shadow:none; padding:0;\">t</abbr></a></li><li class=\"nv-edit\"><a class=\"external text\" href=\"//en.wikipedia.org/w/index.php?title=Template:Healthcare_in_Europe&amp;action=edit\"><abbr title=\"Edit this template\" style=\";;background:none transparent;border:none;-moz-box-shadow:none;-webkit-box-shadow:none;box-shadow:none; padding:0;\">e</abbr></a></li></ul></div><div id=\"Healthcare_in_Europe\" style=\"font-size:114%;margin:0 4em\"><a class=\"mw-selflink selflink\">Healthcare in Europe </a></div></th></tr><tr><th scope=\"row\" class=\"navbox-group\" style=\"width:1%\">Sovereign states</th><td class=\"navbox-list navbox-odd hlist\" style=\"text-align:left;border-left-width:2px;border-left-style:solid;width:100%;padding:0px\"><div style=\"padding:0em 0.25em\">']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lst"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will scrape the first table, and therefore use index 0 in <font color=\"red\">lst</font> to capture the first table name. Now it is about time to read in the table using Beautiful Soup's <font color=\"blue\">find</font> function. A simple option is to type in the table name. You simply select the name in <font color=\"red\">lst</font>, which in this case is \"wikitable sortable\":"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "table=HCE.find('table', {'class', 'wikitable sortable'})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Alternatively, there is a way to automate this step, by capturing the first data from the list, and then stripping off the unneeded characters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "x=lst[0]\n",
    "extr=re.findall('\"([^\"]*)\"', x)\n",
    "table=HCE.find('table', {'class', str(extr).strip(\"'[]'\")})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bs4.element.Tag"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(table)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now it would be good to read in separately the header and row names, so we later can easily make a DataFrame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "headers= [header.text for header in table.find_all('th')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[u'Country\\n',\n",
       " u'Ranking\\n',\n",
       " u'Total Score\\n',\n",
       " u'Patient rights & Information\\n',\n",
       " u'Accessibility (waiting times)\\n',\n",
       " u'Outcomes\\n',\n",
       " u'Range and reach of services provided\\n',\n",
       " u'Prevention\\n',\n",
       " u'Pharmaceuticals\\n']"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "headers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "rows = []\n",
    "for row in table.find_all('tr'):\n",
    "    rows.append([val.text.encode('utf8') for val in row.find_all('td')])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now all elements, -- rows and headers --, are available to build the DataFrame, which we will call <font color=\"red\">df1</font>. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1 = pd.DataFrame(rows, columns=headers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Country</th>\n",
       "      <th>Ranking</th>\n",
       "      <th>Total Score</th>\n",
       "      <th>Patient rights &amp; Information</th>\n",
       "      <th>Accessibility (waiting times)</th>\n",
       "      <th>Outcomes</th>\n",
       "      <th>Range and reach of services provided</th>\n",
       "      <th>Prevention</th>\n",
       "      <th>Pharmaceuticals</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Netherlands\\n</td>\n",
       "      <td>1\\n</td>\n",
       "      <td>924\\n</td>\n",
       "      <td>125\\n</td>\n",
       "      <td>200\\n</td>\n",
       "      <td>278\\n</td>\n",
       "      <td>125\\n</td>\n",
       "      <td>107\\n</td>\n",
       "      <td>89\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Switzerland\\n</td>\n",
       "      <td>2\\n</td>\n",
       "      <td>898\\n</td>\n",
       "      <td>117\\n</td>\n",
       "      <td>225\\n</td>\n",
       "      <td>278\\n</td>\n",
       "      <td>94\\n</td>\n",
       "      <td>101\\n</td>\n",
       "      <td>83\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Denmark\\n</td>\n",
       "      <td>3\\n</td>\n",
       "      <td>864\\n</td>\n",
       "      <td>117\\n</td>\n",
       "      <td>118\\n</td>\n",
       "      <td>267\\n</td>\n",
       "      <td>120\\n</td>\n",
       "      <td>95\\n</td>\n",
       "      <td>119\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Norway\\n</td>\n",
       "      <td>4\\n</td>\n",
       "      <td>850\\n</td>\n",
       "      <td>125\\n</td>\n",
       "      <td>125\\n</td>\n",
       "      <td>289\\n</td>\n",
       "      <td>115\\n</td>\n",
       "      <td>119\\n</td>\n",
       "      <td>78\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Luxembourg\\n</td>\n",
       "      <td>5\\n</td>\n",
       "      <td>850\\n</td>\n",
       "      <td>104\\n</td>\n",
       "      <td>213\\n</td>\n",
       "      <td>244\\n</td>\n",
       "      <td>109\\n</td>\n",
       "      <td>107\\n</td>\n",
       "      <td>72\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Finland\\n</td>\n",
       "      <td>6\\n</td>\n",
       "      <td>856\\n</td>\n",
       "      <td>108\\n</td>\n",
       "      <td>150\\n</td>\n",
       "      <td>289\\n</td>\n",
       "      <td>120\\n</td>\n",
       "      <td>101\\n</td>\n",
       "      <td>78\\n</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Country\\n Ranking\\n Total Score\\n Patient rights & Information\\n  \\\n",
       "0             None      None          None                           None   \n",
       "1    Netherlands\\n       1\\n         924\\n                          125\\n   \n",
       "2    Switzerland\\n       2\\n         898\\n                          117\\n   \n",
       "3        Denmark\\n       3\\n         864\\n                          117\\n   \n",
       "4         Norway\\n       4\\n         850\\n                          125\\n   \n",
       "5     Luxembourg\\n       5\\n         850\\n                          104\\n   \n",
       "6        Finland\\n       6\\n         856\\n                          108\\n   \n",
       "\n",
       "  Accessibility (waiting times)\\n Outcomes\\n  \\\n",
       "0                            None       None   \n",
       "1                           200\\n      278\\n   \n",
       "2                           225\\n      278\\n   \n",
       "3                           118\\n      267\\n   \n",
       "4                           125\\n      289\\n   \n",
       "5                           213\\n      244\\n   \n",
       "6                           150\\n      289\\n   \n",
       "\n",
       "  Range and reach of services provided\\n Prevention\\n Pharmaceuticals\\n  \n",
       "0                                   None         None              None  \n",
       "1                                  125\\n        107\\n              89\\n  \n",
       "2                                   94\\n        101\\n              83\\n  \n",
       "3                                  120\\n         95\\n             119\\n  \n",
       "4                                  115\\n        119\\n              78\\n  \n",
       "5                                  109\\n        107\\n              72\\n  \n",
       "6                                  120\\n        101\\n              78\\n  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head(7)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This table still needs a good amount of preprocessing, on which we will return later."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Health Expenditure**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Of course, other data sources can be scraped as well. So let us load data about health expenditure<font color=\"gray\"><sup>7</sup></font>. These are data per capita, which means that expenditure was corrected for the number of habitants in a country. Saving time, we will now put the script in one code block. This should lead to a DataFrame <font color=\"red\">df2</font>:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<table class=\"wikitable\">\n"
     ]
    }
   ],
   "source": [
    "url = 'https://en.wikipedia.org/wiki/List_of_countries_by_total_health_expenditure_per_capita' \n",
    "r = requests.get(url)\n",
    "HEE = BeautifulSoup(r.content)\n",
    "htmlpage = urllib2.urlopen(url)\n",
    "lst = []\n",
    "for line in htmlpage:\n",
    "    line = line.rstrip()\n",
    "    if re.search('table class', line) :\n",
    "        lst.append(line)\n",
    "x=lst[1]\n",
    "print x\n",
    "extr=re.findall('\"([^\"]*)\"', x)\n",
    "table=HEE.find('table', {'class', 'wikitable sortable'})\n",
    "headers= [header.text for header in table.find_all('th')]\n",
    "rows = []\n",
    "for row in table.find_all('tr'):\n",
    "    rows.append([val.text.encode('utf8') for val in row.find_all('td')])\n",
    "headers = [i.replace(\"\\n\", \"\") for i in headers]\n",
    "df2 = pd.DataFrame(rows, columns=headers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Country</th>\n",
       "      <th>2013</th>\n",
       "      <th>2014</th>\n",
       "      <th>2015</th>\n",
       "      <th>2016</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Australia</td>\n",
       "      <td>4,186</td>\n",
       "      <td>4,289</td>\n",
       "      <td>4,493</td>\n",
       "      <td>4,708\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Austria</td>\n",
       "      <td>4,861</td>\n",
       "      <td>5,001</td>\n",
       "      <td>5,100</td>\n",
       "      <td>5,227\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Belgium</td>\n",
       "      <td>4,505</td>\n",
       "      <td>4,656</td>\n",
       "      <td>4,778</td>\n",
       "      <td>4,840\\n</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Canada</td>\n",
       "      <td>4,461</td>\n",
       "      <td>4,502</td>\n",
       "      <td>4,613</td>\n",
       "      <td>4,753\\n</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Country   2013   2014   2015     2016\n",
       "0       None   None   None   None     None\n",
       "1  Australia  4,186  4,289  4,493  4,708\\n\n",
       "2    Austria  4,861  5,001  5,100  5,227\\n\n",
       "3    Belgium  4,505  4,656  4,778  4,840\\n\n",
       "4     Canada  4,461  4,502  4,613  4,753\\n"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "## **Additional preprocessing steps**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we look at the dataframes, we see that there are still some issues that prohibit numeric computations. There are undesired characters ('\\n'), undesired decimal format (comma should be removed), there are cells with non-numeric characters ('x') that should be NAN, and several columns should be numeric instead of objects. To resolve this, we will write a preprocessing function called **preproc**, which accepts as input any DataFrame (in our case df1 and df2). Please note that each webtable may need its unique collection of preprocessing steps.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def preproc(dat):\n",
    "    dat.dropna(axis=0, how='all', inplace=True)\n",
    "    dat.columns = dat.columns.str.replace(\"\\n\", \"\")    \n",
    "    dat.replace([\"\\n\"], [\"\"], regex=True, inplace=True)\n",
    "    dat.replace([\",\"], [\"\"], regex=True, inplace=True)\n",
    "    dat.replace(r\"\\b[a-zA-Z]\\b\", np.nan, regex=True, inplace=True)\n",
    "    dat = dat.apply(pd.to_numeric, errors='ignore')    \n",
    "    return(dat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1 = preproc(df1)\n",
    "df2 = preproc(df2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Apparently, after this preprocessing there are some NANs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "8\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "print(df1.isnull().sum().sum())\n",
    "print(df2.isnull().sum().sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we display where the NANs occur. In fact, when we check the original table, we can see that Cyprus has values \"x\", which were in our preproc function changed to NANs ( https://en.wikipedia.org/wiki/Healthcare_in_Europe )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Country</th>\n",
       "      <th>Ranking</th>\n",
       "      <th>Total Score</th>\n",
       "      <th>Patient rights &amp; Information</th>\n",
       "      <th>Accessibility (waiting times)</th>\n",
       "      <th>Outcomes</th>\n",
       "      <th>Range and reach of services provided</th>\n",
       "      <th>Prevention</th>\n",
       "      <th>Pharmaceuticals</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>Cyprus</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Country  Ranking  Total Score  Patient rights & Information  \\\n",
       "35   Cyprus      NaN          NaN                           NaN   \n",
       "\n",
       "    Accessibility (waiting times)  Outcomes  \\\n",
       "35                            NaN       NaN   \n",
       "\n",
       "    Range and reach of services provided  Prevention  Pharmaceuticals  \n",
       "35                                   NaN         NaN              NaN  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1[df1.isnull().any(axis=1)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we remove the NANs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1.dropna(axis=0, how='any', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "At this point we inspect the data types:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Country                                  object\n",
       "Ranking                                 float64\n",
       "Total Score                             float64\n",
       "Patient rights & Information            float64\n",
       "Accessibility (waiting times)           float64\n",
       "Outcomes                                float64\n",
       "Range and reach of services provided    float64\n",
       "Prevention                              float64\n",
       "Pharmaceuticals                         float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Country    object\n",
       "2013        int64\n",
       "2014        int64\n",
       "2015        int64\n",
       "2016        int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The columnnames are a bit long, and so it would be good to use shorter names."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1.columns = ['Country', 'Ranking', 'totalscore', 'Patientrights', 'Accessibility', 'Outcomes', 'Range', 'Prevention', 'Pharmaceuticals']\n",
    "df2.columns = ['Country', 'y2013', 'y2014', 'y2015', 'y2016']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "\n",
    "## **Merging different data**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "It should be clear from this example that webscraping can be important to quickly grasp data.\n",
    "Webscraping may be particularly useful when you need to automate data processing:\n",
    "\n",
    "* Webdata change regularly and need to be stored repeatedly.\n",
    "\n",
    "* A large number of data sources, for example tables, need to be loaded and merged.\n",
    "\n",
    "Let us elaborate the last point a bit more. If the two tables that we just scraped need to be merged, this can be done in Python. For example, if we want to merge on the column \"Country\", we would use the following code (we use the <font color=\"blue\">.head()</font> function to limit the output)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Country</th>\n",
       "      <th>Ranking</th>\n",
       "      <th>totalscore</th>\n",
       "      <th>Patientrights</th>\n",
       "      <th>Accessibility</th>\n",
       "      <th>Outcomes</th>\n",
       "      <th>Range</th>\n",
       "      <th>Prevention</th>\n",
       "      <th>Pharmaceuticals</th>\n",
       "      <th>y2013</th>\n",
       "      <th>y2014</th>\n",
       "      <th>y2015</th>\n",
       "      <th>y2016</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Netherlands</td>\n",
       "      <td>1.0</td>\n",
       "      <td>924.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>278.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Switzerland</td>\n",
       "      <td>2.0</td>\n",
       "      <td>898.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>225.0</td>\n",
       "      <td>278.0</td>\n",
       "      <td>94.0</td>\n",
       "      <td>101.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Denmark</td>\n",
       "      <td>3.0</td>\n",
       "      <td>864.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>267.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>119.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Norway</td>\n",
       "      <td>4.0</td>\n",
       "      <td>850.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>289.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>119.0</td>\n",
       "      <td>78.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Luxembourg</td>\n",
       "      <td>5.0</td>\n",
       "      <td>850.0</td>\n",
       "      <td>104.0</td>\n",
       "      <td>213.0</td>\n",
       "      <td>244.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>72.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Country  Ranking  totalscore  Patientrights  Accessibility  Outcomes  \\\n",
       "0    Netherlands      1.0       924.0          125.0          200.0     278.0   \n",
       "1    Switzerland      2.0       898.0          117.0          225.0     278.0   \n",
       "2        Denmark      3.0       864.0          117.0          118.0     267.0   \n",
       "3         Norway      4.0       850.0          125.0          125.0     289.0   \n",
       "4     Luxembourg      5.0       850.0          104.0          213.0     244.0   \n",
       "\n",
       "   Range  Prevention  Pharmaceuticals  y2013  y2014  y2015  y2016  \n",
       "0  125.0       107.0             89.0    NaN    NaN    NaN    NaN  \n",
       "1   94.0       101.0             83.0    NaN    NaN    NaN    NaN  \n",
       "2  120.0        95.0            119.0    NaN    NaN    NaN    NaN  \n",
       "3  115.0       119.0             78.0    NaN    NaN    NaN    NaN  \n",
       "4  109.0       107.0             72.0    NaN    NaN    NaN    NaN  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.merge(df1, df2, how='left', on='Country').head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The health expenditure data give unexpected NaNs (missing values). Something must be wrong here. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we need to inspect this in more detail. Normally using the <font color=\"blue\">set</font> function to check for overlap between the countries in the two DataFrames should give a set. But now it is empty. This may suggest that the tables cannot merge because the <font color=\"red\">Country</font> names are not identical."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "set()"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(df1['Country']) & set(df2['Country'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's inspect these tables more closely. Something must be wrong because there are several countries that are overlapping and there are no spelling errors. If we write the table to a csv file to check the raw text we may discover hidden symbols."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1.to_csv('df1example.csv', sep=\",\")\n",
    "df2.to_csv('df2example.csv', sep=\",\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you now open the csv file in a spreadsheet software you would discover that there are some hidden characters in the <font color=\"red\">df1example.csv</font> file. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using the <font color=\"blue\">repr</font> function we can reveal the hidden characters<sup>8</sup>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1        \\xc2\\xa0Netherlands\\n2       \\xc2\\xa0\\xc2\\xa0Switzerland\\n3            \\xc2\\xa0Denmark\\n4             \\xc2\\xa0Norway\\n5         \\xc2\\xa0Luxembourg\\n6            \\xc2\\xa0Finland\\n7            \\xc2\\xa0Germany\\n8            \\xc2\\xa0Belgium\\n9            \\xc2\\xa0Iceland\\n10            \\xc2\\xa0France\\n11           \\xc2\\xa0Austria\\n12            \\xc2\\xa0Sweden\\n13          \\xc2\\xa0Slovakia\\n14          \\xc2\\xa0Portugal\\n15    \\xc2\\xa0United Kingdom\\n16          \\xc2\\xa0Slovenia\\n17           \\xc2\\xa0Czechia\\n18             \\xc2\\xa0Spain\\n19           \\xc2\\xa0Estonia\\n20            \\xc2\\xa0Serbia\\n21             \\xc2\\xa0Italy\\n22         \\xc2\\xa0Macedonia\\n23             \\xc2\\xa0Malta\\n24           \\xc2\\xa0Ireland\\n25        \\xc2\\xa0Montenegro\\n26           \\xc2\\xa0Croatia\\n27           \\xc2\\xa0Albania\\n28            \\xc2\\xa0Latvia\\n29            \\xc2\\xa0Poland\\n30           \\xc2\\xa0Hungary\\n31         \\xc2\\xa0Lithuania\\n32            \\xc2\\xa0Greece\\n33          \\xc2\\xa0Bulgaria\\n34           \\xc2\\xa0Romania\\nName: Country, dtype: object'"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "repr(df1['Country'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1           Australia\\n2             Austria\\n3             Belgium\\n4              Canada\\n5               Chile\\n6      Czech Republic\\n7             Denmark\\n8             Estonia\\n9             Finland\\n10             France\\n11            Germany\\n12             Greece\\n13            Hungary\\n14            Iceland\\n15            Ireland\\n16             Israel\\n17              Italy\\n18              Japan\\n19              Korea\\n20             Latvia\\n21         Luxembourg\\n22             Mexico\\n23        Netherlands\\n24        New Zealand\\n25             Norway\\n26             Poland\\n27           Portugal\\n28    Slovak Republic\\n29           Slovenia\\n30              Spain\\n31             Sweden\\n32        Switzerland\\n33             Turkey\\n34     United Kingdom\\n35      United States\\nName: Country, dtype: object'"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "repr(df2['Country'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Luckily, all unwanted hidden characters are identical across the countries. That is, the characters are \"\\xc2\\xa0\" and \"\\n\". Using the <font color=\"blue\">replace</font> function we could remove this part."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Country</th>\n",
       "      <th>Ranking</th>\n",
       "      <th>totalscore</th>\n",
       "      <th>Patientrights</th>\n",
       "      <th>Accessibility</th>\n",
       "      <th>Outcomes</th>\n",
       "      <th>Range</th>\n",
       "      <th>Prevention</th>\n",
       "      <th>Pharmaceuticals</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Netherlands</td>\n",
       "      <td>1.0</td>\n",
       "      <td>924.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>278.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>89.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Switzerland</td>\n",
       "      <td>2.0</td>\n",
       "      <td>898.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>225.0</td>\n",
       "      <td>278.0</td>\n",
       "      <td>94.0</td>\n",
       "      <td>101.0</td>\n",
       "      <td>83.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Denmark</td>\n",
       "      <td>3.0</td>\n",
       "      <td>864.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>267.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>119.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Norway</td>\n",
       "      <td>4.0</td>\n",
       "      <td>850.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>289.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>119.0</td>\n",
       "      <td>78.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Luxembourg</td>\n",
       "      <td>5.0</td>\n",
       "      <td>850.0</td>\n",
       "      <td>104.0</td>\n",
       "      <td>213.0</td>\n",
       "      <td>244.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>72.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Country  Ranking  totalscore  Patientrights  Accessibility  Outcomes  \\\n",
       "1    Netherlands      1.0       924.0          125.0          200.0     278.0   \n",
       "2    Switzerland      2.0       898.0          117.0          225.0     278.0   \n",
       "3        Denmark      3.0       864.0          117.0          118.0     267.0   \n",
       "4         Norway      4.0       850.0          125.0          125.0     289.0   \n",
       "5     Luxembourg      5.0       850.0          104.0          213.0     244.0   \n",
       "\n",
       "   Range  Prevention  Pharmaceuticals  \n",
       "1  125.0       107.0             89.0  \n",
       "2   94.0       101.0             83.0  \n",
       "3  120.0        95.0            119.0  \n",
       "4  115.0       119.0             78.0  \n",
       "5  109.0       107.0             72.0  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df1.replace([\"\\n\"],[\"\"], regex=True, inplace=True)\n",
    "df1.replace([\"\\xc2\\xa0\"], [\"\"], regex=True, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1        Netherlands\\n2        Switzerland\\n3            Denmark\\n4             Norway\\n5         Luxembourg\\n6            Finland\\n7            Germany\\n8            Belgium\\n9            Iceland\\n10            France\\n11           Austria\\n12            Sweden\\n13          Slovakia\\n14          Portugal\\n15    United Kingdom\\n16          Slovenia\\n17           Czechia\\n18             Spain\\n19           Estonia\\n20            Serbia\\n21             Italy\\n22         Macedonia\\n23             Malta\\n24           Ireland\\n25        Montenegro\\n26           Croatia\\n27           Albania\\n28            Latvia\\n29            Poland\\n30           Hungary\\n31         Lithuania\\n32            Greece\\n33          Bulgaria\\n34           Romania\\nName: Country, dtype: object'"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "repr(df1['Country'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Country</th>\n",
       "      <th>Ranking</th>\n",
       "      <th>totalscore</th>\n",
       "      <th>Patientrights</th>\n",
       "      <th>Accessibility</th>\n",
       "      <th>Outcomes</th>\n",
       "      <th>Range</th>\n",
       "      <th>Prevention</th>\n",
       "      <th>Pharmaceuticals</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Netherlands</td>\n",
       "      <td>1.0</td>\n",
       "      <td>924.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>200.0</td>\n",
       "      <td>278.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>89.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Switzerland</td>\n",
       "      <td>2.0</td>\n",
       "      <td>898.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>225.0</td>\n",
       "      <td>278.0</td>\n",
       "      <td>94.0</td>\n",
       "      <td>101.0</td>\n",
       "      <td>83.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Denmark</td>\n",
       "      <td>3.0</td>\n",
       "      <td>864.0</td>\n",
       "      <td>117.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>267.0</td>\n",
       "      <td>120.0</td>\n",
       "      <td>95.0</td>\n",
       "      <td>119.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Norway</td>\n",
       "      <td>4.0</td>\n",
       "      <td>850.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>125.0</td>\n",
       "      <td>289.0</td>\n",
       "      <td>115.0</td>\n",
       "      <td>119.0</td>\n",
       "      <td>78.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Luxembourg</td>\n",
       "      <td>5.0</td>\n",
       "      <td>850.0</td>\n",
       "      <td>104.0</td>\n",
       "      <td>213.0</td>\n",
       "      <td>244.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>72.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Country  Ranking  totalscore  Patientrights  Accessibility  Outcomes  \\\n",
       "1  Netherlands      1.0       924.0          125.0          200.0     278.0   \n",
       "2  Switzerland      2.0       898.0          117.0          225.0     278.0   \n",
       "3      Denmark      3.0       864.0          117.0          118.0     267.0   \n",
       "4       Norway      4.0       850.0          125.0          125.0     289.0   \n",
       "5   Luxembourg      5.0       850.0          104.0          213.0     244.0   \n",
       "\n",
       "   Range  Prevention  Pharmaceuticals  \n",
       "1  125.0       107.0             89.0  \n",
       "2   94.0       101.0             83.0  \n",
       "3  120.0        95.0            119.0  \n",
       "4  115.0       119.0             78.0  \n",
       "5  109.0       107.0             72.0  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Alright. To inspect if it worked, we could use <font color=\"blue\">repr</font> function again and should now see that the undesired characters are removed. Using the <font color=\"blue\">set</font> function again we can see now that there are quite some countries that overlap:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Austria',\n",
       " 'Belgium',\n",
       " 'Denmark',\n",
       " 'Estonia',\n",
       " 'Finland',\n",
       " 'France',\n",
       " 'Germany',\n",
       " 'Greece',\n",
       " 'Hungary',\n",
       " 'Iceland',\n",
       " 'Ireland',\n",
       " 'Italy',\n",
       " 'Latvia',\n",
       " 'Luxembourg',\n",
       " 'Netherlands',\n",
       " 'Norway',\n",
       " 'Poland',\n",
       " 'Portugal',\n",
       " 'Slovenia',\n",
       " 'Spain',\n",
       " 'Sweden',\n",
       " 'Switzerland',\n",
       " 'United Kingdom'}"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(df1['Country']) & set(df2['Country'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now it is about time to merge the two tables. A few countries have NaNs but in these cases for the expenditure data there were no values. We make a new DataFrame <font color=\"red\">df3</font> from the merged data. After that we drop any rows where there are NaNs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df3 = pd.merge(df1,df2, how='left', on='Country')\n",
    "df3.dropna(how='any', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(23, 13)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "\n",
    "## **Data visualization**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "We managed to bring the data together. The data are ready for analyses and visualization. \n",
    "We could visually inspect the relation between different variables using a scatterplot. The package **adjustText** takes care that the country labels do not overlap<font color=\"gray\"><sup>9</sup></font>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "reload(sys)\n",
    "sys.setdefaultencoding('utf8')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "For the visualization, we will compute the average across years."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df3['Avg'] = df3.loc[:, ['y2013', 'y2014', 'y2015', 'y2016'] ].mean(axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We tune the axes and prepare the minimum and maximum values on the X- and Y-axis:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "xmin=df3.Patientrights.min()-10\n",
    "xmax=df3.Patientrights.max()+10\n",
    "ymin=df3.Outcomes.min()-10\n",
    "ymax=df3.Outcomes.max()+10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA00AAAM9CAYAAABAIZjtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f4ea0307d50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "from adjustText import adjust_text\n",
    "\n",
    "def plot_df3(adjust=True):\n",
    "    mpl.rcParams['font.size'] = 12.0\n",
    "    plt.figure(figsize = (14, 14))\n",
    "    plt.scatter(df3.Patientrights, df3.Outcomes, facecolors='none', edgecolors='red', linewidth=1.2, s=1*df3.Avg)\n",
    "    texts = []\n",
    "    plt.title('Relation between different health parameters')\n",
    "    plt.xlabel('Patient rights')\n",
    "    plt.ylabel('Outcomes')\n",
    "    plt.xlim(xmin, xmax)\n",
    "    plt.ylim(ymin, ymax)\n",
    "\n",
    "    for x, y, s in zip(df3['Patientrights'], df3['Outcomes'], df3['Country']):\n",
    "        texts.append(plt.text(x, y, s, size=12))\n",
    "    if adjust:\n",
    "        plt.title(str(adjust_text(texts, arrowprops=dict(arrowstyle=\"-\", color='black', lw=0.5))\n",
    "))\n",
    "\n",
    "_ = plot_df3()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The visualization suggests a relation between patient rights and outcomes. The diameter of the circles displays expenditure. The higher expenditure countries have better patient rights and outcomes (mostly in the right top corner)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **Summary**\n",
    "\n",
    "The current blog showed how webtables can be scraped by using Python.\n",
    "A certain amount of preprocessing is necessary before these data can be visualized.\n",
    "Note that the world of webscraping is dynamic. The used websites and tables may be updated and therefore may require regular updates of Python code. \n",
    "\n",
    "Any questions or comments? \n",
    "\n",
    "Please feel free to contact me : \n",
    "\n",
    "Ruthger Righart\n",
    "\n",
    "E: rrighart@googlemail.com\n",
    "\n",
    "W: https://www.rrighart.com\n",
    "___\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## **References**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "[1]. Data is the new oil. https://www.changethislimited.co.uk/2017/01/data-is-the-new-oil/\n",
    "\n",
    "[2]. Data on internet. https://www.livescience.com/54094-how-big-is-the-internet.html\n",
    "\n",
    "[3]. Beautiful Soup. https://www.crummy.com/software/BeautifulSoup/bs4/\n",
    "\n",
    "[4]. Healthcare Europe. https://en.wikipedia.org/wiki/Healthcare_in_Europe\n",
    "\n",
    "[5]. Visualizing European healthcare using Tableau. https://rrighart.github.io/HE-Tableau/\n",
    "\n",
    "[6]. Scraping tables. https://stackoverflow.com/questions/17196018/extracting-table-contents-from-html-with-python-and-beautifulsoup\n",
    "\n",
    "[7]. Health expenditure. https://en.wikipedia.org/wiki/List_of_countries_by_total_health_expenditure_per_capita\n",
    "\n",
    "[8]. Hidden characters. https://stackoverflow.com/questions/31341351/how-can-i-identify-invisible-characters-in-python-strings\n",
    "\n",
    "[9]. Adjust text package. https://github.com/Phlya/adjustText/blob/master/examples/Examples.ipynb\n",
    "\n",
    "___\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "(c) 2018 . https://www.rrighart.com | https://rrighart.github.io"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}